Re: [PATCH] Fix for PR52009 - Another missed tail merging opportunity
On Wed, Jul 4, 2012 at 8:07 PM, Tom de Vries tom_devr...@mentor.com wrote:
On 31/01/12 22:07, Tom de Vries wrote:
On 31/01/12 22:05, Tom de Vries wrote:
Richard,
Sorry, with patch this time.
this patch fixes PR52009.
Consider this test-case:
...
int z;
void
foo (int y)
{
if (y == 6)
z = 5;
else
z = 5;
}
...
Currently, compiling with -O2 gives us this representation at
pr51879-7.c.094t.pre:
...
# BLOCK 3 freq:1991
# PRED: 2 [19.9%] (true,exec)
# .MEMD.1710_4 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
goto bb 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 4 freq:8009
# PRED: 2 [80.1%] (false,exec)
# .MEMD.1710_5 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 5 freq:1
# PRED: 3 [100.0%] (fallthru,exec) 4 [100.0%] (fallthru,exec)
# .MEMD.1710_2 = PHI .MEMD.1710_4(3), .MEMD.1710_5(4)
# VUSE .MEMD.1710_2
return;
...
Blocks 3 and 4 are not tail-merged.
The patch allows the example to be tail-merged by:
- value numbering .MEMD.1710_4 and .MEMD.1710_5 equal
- comparing gimple_vdef value numbers for assignments during tail-merge
Bootstrapped and reg-tested on x86_64.
OK for stage1?
Richard,
I did some trivial changes such that the patch is applicable again to trunk,
and
added a test-case, which you suggested in
http://gcc.gnu.org/ml/gcc-patches/2012-05/msg00220.html.
The test-case handles the case of 2 identical calls assigning to different
deferenced pointers, where the pointers are value-numbered the same:
...
void bar (int c, int *p)
{
int *q = p;
if (c)
*p = foo ();
else
*q = foo ();
}
...
The representation before pre looks like this:
...
# BLOCK 3 freq:3900
# PRED: 2 [39.0%] (true,exec)
# .MEMD.1724_8 = VDEF .MEMD.1724_7(D)
# USE = nonlocal
# CLB = nonlocal
D.1721_4 = fooD.1712 ();
# .MEMD.1724_9 = VDEF .MEMD.1724_8
*pD.1714_1(D) = D.1721_4;
goto bb 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 4 freq:6100
# PRED: 2 [61.0%] (false,exec)
# .MEMD.1724_10 = VDEF .MEMD.1724_7(D)
# USE = nonlocal
# CLB = nonlocal
D.1723_5 = fooD.1712 ();
# .MEMD.1724_11 = VDEF .MEMD.1724_10
*qD.1717_2 = D.1723_5;
# SUCC: 5 [100.0%] (fallthru,exec)
...
In pre, we're already value numbering the result and vdef of the calls the
same,
and the pointers the same:
...
Value numbers:
q_2 = p_1(D)
D.1723_5 = D.1721_4
.MEM_10 = .MEM_8
...
But not the vdefs of the stores. This patch implements that, and allows the
blocks to be merged.
ok for trunk?
Ok.
Thanks,
Richard.
Thanks,
- Tom
2012-07-04 Tom de Vries t...@codesourcery.com
PR tree-optimization/52009
* tree-ssa-tail-merge.c (gimple_equal_p): For GIMPLE_ASSIGN, compare
value numbers of gimple_vdef.
* tree-ssa-sccvn.h (vn_reference_insert): Add vdef parameter to
prototype.
* tree-ssa-sccvn.c (copy_reference_ops_from_ref): Handle MODIFY_EXPR.
(vn_reference_insert): Add and handle vdef parameter.
(visit_reference_op_load): Add argument to vn_reference_insert call.
(visit_reference_op_store): Find value number of vdef of store.
Insert
value number of vdef of store.
* gcc.dg/pr51879-7.c: New test.
* gcc.dg/pr51879-18.c: New test.
Re: [PATCH] Fix for PR52009 - Another missed tail merging opportunity
On 31/01/12 22:07, Tom de Vries wrote:
On 31/01/12 22:05, Tom de Vries wrote:
Richard,
Sorry, with patch this time.
this patch fixes PR52009.
Consider this test-case:
...
int z;
void
foo (int y)
{
if (y == 6)
z = 5;
else
z = 5;
}
...
Currently, compiling with -O2 gives us this representation at
pr51879-7.c.094t.pre:
...
# BLOCK 3 freq:1991
# PRED: 2 [19.9%] (true,exec)
# .MEMD.1710_4 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
goto bb 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 4 freq:8009
# PRED: 2 [80.1%] (false,exec)
# .MEMD.1710_5 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 5 freq:1
# PRED: 3 [100.0%] (fallthru,exec) 4 [100.0%] (fallthru,exec)
# .MEMD.1710_2 = PHI .MEMD.1710_4(3), .MEMD.1710_5(4)
# VUSE .MEMD.1710_2
return;
...
Blocks 3 and 4 are not tail-merged.
The patch allows the example to be tail-merged by:
- value numbering .MEMD.1710_4 and .MEMD.1710_5 equal
- comparing gimple_vdef value numbers for assignments during tail-merge
Bootstrapped and reg-tested on x86_64.
OK for stage1?
Richard,
I did some trivial changes such that the patch is applicable again to trunk, and
added a test-case, which you suggested in
http://gcc.gnu.org/ml/gcc-patches/2012-05/msg00220.html.
The test-case handles the case of 2 identical calls assigning to different
deferenced pointers, where the pointers are value-numbered the same:
...
void bar (int c, int *p)
{
int *q = p;
if (c)
*p = foo ();
else
*q = foo ();
}
...
The representation before pre looks like this:
...
# BLOCK 3 freq:3900
# PRED: 2 [39.0%] (true,exec)
# .MEMD.1724_8 = VDEF .MEMD.1724_7(D)
# USE = nonlocal
# CLB = nonlocal
D.1721_4 = fooD.1712 ();
# .MEMD.1724_9 = VDEF .MEMD.1724_8
*pD.1714_1(D) = D.1721_4;
goto bb 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 4 freq:6100
# PRED: 2 [61.0%] (false,exec)
# .MEMD.1724_10 = VDEF .MEMD.1724_7(D)
# USE = nonlocal
# CLB = nonlocal
D.1723_5 = fooD.1712 ();
# .MEMD.1724_11 = VDEF .MEMD.1724_10
*qD.1717_2 = D.1723_5;
# SUCC: 5 [100.0%] (fallthru,exec)
...
In pre, we're already value numbering the result and vdef of the calls the same,
and the pointers the same:
...
Value numbers:
q_2 = p_1(D)
D.1723_5 = D.1721_4
.MEM_10 = .MEM_8
...
But not the vdefs of the stores. This patch implements that, and allows the
blocks to be merged.
ok for trunk?
Thanks,
- Tom
2012-07-04 Tom de Vries t...@codesourcery.com
PR tree-optimization/52009
* tree-ssa-tail-merge.c (gimple_equal_p): For GIMPLE_ASSIGN, compare
value numbers of gimple_vdef.
* tree-ssa-sccvn.h (vn_reference_insert): Add vdef parameter to
prototype.
* tree-ssa-sccvn.c (copy_reference_ops_from_ref): Handle MODIFY_EXPR.
(vn_reference_insert): Add and handle vdef parameter.
(visit_reference_op_load): Add argument to vn_reference_insert call.
(visit_reference_op_store): Find value number of vdef of store. Insert
value number of vdef of store.
* gcc.dg/pr51879-7.c: New test.
* gcc.dg/pr51879-18.c: New test.
Index: gcc/tree-ssa-tail-merge.c
===
--- gcc/tree-ssa-tail-merge.c (revision 189007)
+++ gcc/tree-ssa-tail-merge.c (working copy)
@@ -1119,6 +1119,14 @@ gimple_equal_p (same_succ same_succ, gim
case GIMPLE_ASSIGN:
lhs1 = gimple_get_lhs (s1);
lhs2 = gimple_get_lhs (s2);
+ if (gimple_vdef (s1))
+ {
+ if (vn_valueize (gimple_vdef (s1)) != vn_valueize (gimple_vdef (s2)))
+ return false;
+ if (TREE_CODE (lhs1) != SSA_NAME
+ TREE_CODE (lhs2) != SSA_NAME)
+ return true;
+ }
return (TREE_CODE (lhs1) == SSA_NAME
TREE_CODE (lhs2) == SSA_NAME
vn_valueize (lhs1) == vn_valueize (lhs2));
Index: gcc/tree-ssa-sccvn.c
===
--- gcc/tree-ssa-sccvn.c (revision 189007)
+++ gcc/tree-ssa-sccvn.c (working copy)
@@ -628,6 +628,9 @@ copy_reference_ops_from_ref (tree ref, V
switch (temp.opcode)
{
+ case MODIFY_EXPR:
+ temp.op0 = TREE_OPERAND (ref, 1);
+ break;
case WITH_SIZE_EXPR:
temp.op0 = TREE_OPERAND (ref, 1);
temp.off = 0;
@@ -748,6 +751,7 @@ copy_reference_ops_from_ref (tree ref, V
VEC_safe_push (vn_reference_op_s, heap, *result, temp);
if (REFERENCE_CLASS_P (ref)
+ || TREE_CODE (ref) == MODIFY_EXPR
|| TREE_CODE (ref) == WITH_SIZE_EXPR
|| (TREE_CODE (ref) == ADDR_EXPR
!is_gimple_min_invariant (ref)))
@@ -1941,7 +1945,7 @@ vn_reference_lookup (tree op, tree vuse,
RESULT, and return the resulting reference structure we created. */
vn_reference_t
-vn_reference_insert (tree op, tree result, tree vuse)
+vn_reference_insert (tree op, tree result, tree vuse, tree vdef)
{
void **slot;
vn_reference_t vr1;
@@ -1957,6 +1961,7 @@
Re: [PATCH] Fix for PR52009 - Another missed tail merging opportunity
On 31/01/12 22:05, Tom de Vries wrote:
Richard,
Sorry, with patch this time.
this patch fixes PR52009.
Consider this test-case:
...
int z;
void
foo (int y)
{
if (y == 6)
z = 5;
else
z = 5;
}
...
Currently, compiling with -O2 gives us this representation at
pr51879-7.c.094t.pre:
...
# BLOCK 3 freq:1991
# PRED: 2 [19.9%] (true,exec)
# .MEMD.1710_4 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
goto bb 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 4 freq:8009
# PRED: 2 [80.1%] (false,exec)
# .MEMD.1710_5 = VDEF .MEMD.1710_3(D)
zD.1702 = 5;
# SUCC: 5 [100.0%] (fallthru,exec)
# BLOCK 5 freq:1
# PRED: 3 [100.0%] (fallthru,exec) 4 [100.0%] (fallthru,exec)
# .MEMD.1710_2 = PHI .MEMD.1710_4(3), .MEMD.1710_5(4)
# VUSE .MEMD.1710_2
return;
...
Blocks 3 and 4 are not tail-merged.
The patch allows the example to be tail-merged by:
- value numbering .MEMD.1710_4 and .MEMD.1710_5 equal
- comparing gimple_vdef value numbers for assignments during tail-merge
Bootstrapped and reg-tested on x86_64.
OK for stage1?
Thanks,
- Tom
2012-01-31 Tom de Vries t...@codesourcery.com
PR tree-optimization/52009
* tree-ssa-tail-merge.c (gimple_equal_p): For GIMPLE_ASSIGN, compare
value numbers of gimple_vdef.
* tree-ssa-sccvn.h (struct vn_reference_s): Add vdef field.
(vn_reference_insert): Add vdef parameter to prototype.
* tree-ssa-sccvn.c (copy_reference_ops_from_ref): Handle MODIFY_EXPR.
(vn_reference_insert): Add and handle vdef parameter.
(visit_reference_op_load): Add argument to vn_reference_insert call.
(visit_reference_op_store): Find value number of vdef of store. Insert
value number of vdef of store.
* gcc.dg/pr51879-7.c: New test.
Index: gcc/tree-ssa-tail-merge.c
===
--- gcc/tree-ssa-tail-merge.c (revision 183325)
+++ gcc/tree-ssa-tail-merge.c (working copy)
@@ -1087,6 +1087,14 @@ gimple_equal_p (same_succ same_succ, gim
case GIMPLE_ASSIGN:
lhs1 = gimple_get_lhs (s1);
lhs2 = gimple_get_lhs (s2);
+ if (gimple_vdef (s1))
+ {
+ if (vn_valueize (gimple_vdef (s1)) != vn_valueize (gimple_vdef (s2)))
+ return false;
+ if (TREE_CODE (lhs1) != SSA_NAME
+ TREE_CODE (lhs2) != SSA_NAME)
+ return true;
+ }
return (TREE_CODE (lhs1) == SSA_NAME
TREE_CODE (lhs2) == SSA_NAME
vn_valueize (lhs1) == vn_valueize (lhs2));
Index: gcc/tree-ssa-sccvn.c
===
--- gcc/tree-ssa-sccvn.c (revision 183325)
+++ gcc/tree-ssa-sccvn.c (working copy)
@@ -624,6 +624,9 @@ copy_reference_ops_from_ref (tree ref, V
switch (temp.opcode)
{
+ case MODIFY_EXPR:
+ temp.op0 = TREE_OPERAND (ref, 1);
+ break;
case MEM_REF:
/* The base address gets its own vn_reference_op_s structure. */
temp.op0 = TREE_OPERAND (ref, 1);
@@ -740,6 +743,7 @@ copy_reference_ops_from_ref (tree ref, V
VEC_safe_push (vn_reference_op_s, heap, *result, temp);
if (REFERENCE_CLASS_P (ref)
+ || TREE_CODE (ref) == MODIFY_EXPR
|| (TREE_CODE (ref) == ADDR_EXPR
!is_gimple_min_invariant (ref)))
ref = TREE_OPERAND (ref, 0);
@@ -1928,7 +1932,7 @@ vn_reference_lookup (tree op, tree vuse,
RESULT, and return the resulting reference structure we created. */
vn_reference_t
-vn_reference_insert (tree op, tree result, tree vuse)
+vn_reference_insert (tree op, tree result, tree vuse, tree vdef)
{
void **slot;
vn_reference_t vr1;
@@ -1944,6 +1948,7 @@ vn_reference_insert (tree op, tree resul
vr1-set = get_alias_set (op);
vr1-hashcode = vn_reference_compute_hash (vr1);
vr1-result = TREE_CODE (result) == SSA_NAME ? SSA_VAL (result) : result;
+ vr1-vdef = vdef;
slot = htab_find_slot_with_hash (current_info-references, vr1, vr1-hashcode,
INSERT);
@@ -2725,7 +2730,7 @@ visit_reference_op_load (tree lhs, tree
else
{
changed = set_ssa_val_to (lhs, lhs);
- vn_reference_insert (op, lhs, last_vuse);
+ vn_reference_insert (op, lhs, last_vuse, NULL_TREE);
}
return changed;
@@ -2739,8 +2744,11 @@ static bool
visit_reference_op_store (tree lhs, tree op, gimple stmt)
{
bool changed = false;
- tree result;
+ vn_reference_t vnresult = NULL;
+ tree result, assign;
bool resultsame = false;
+ tree vuse = gimple_vuse (stmt);
+ tree vdef = gimple_vdef (stmt);
/* First we want to lookup using the *vuses* from the store and see
if there the last store to this location with the same address
@@ -2758,7 +2766,7 @@ visit_reference_op_store (tree lhs, tree
Otherwise, the vdefs for the store are used when inserting into
the table, since the store generates a new memory state. */
- result = vn_reference_lookup (lhs, gimple_vuse (stmt), VN_NOWALK, NULL);
+ result = vn_reference_lookup (lhs,
